Operant Conditioning is one of the most significant contributions to learning science of the past 75 years. It was developed by B. F. Skinner and consists of four distinct parts which work separately as well as together. These four parts of the Operant Conditioning Learning Model are explained individually in the articles that follow. A great writer by the name of Dr. Martyn Lloyd Jones once said, “There is nothing more likely to lead to error than to begin with the part rather than the whole.”

To emphasize one part of this learning model and neglect other parts will result is a frustrated dog owner and a confuse dog.  Sixty years ago punishment was emphasized and we had obedient dogs, but they were not very happy. Today we have emphasized “positive reinforcement” and as a result we have very happy dogs, but they are not for obedient. You can have both if you understand the balance which exists in the learning principles of Operant Conditioning

Positive Reinforcement

Positive Reinforcement is one segment of a four part learning model called Operant Conditioning, developed by B.F. Skinner in 1956. Operant Conditioning is one type of associative learning. This learning theory is based on the assumption that animals, as well as people, learn that their behavior earns consequences. The next time you see a cute puppy sitting obediently, looking at you with his head cocked and wonder what he is thinking, consider that the puppy may be wanting to ask you, “How does my behavior affect what happens to me?”

Although Operant Conditioning is a four part learning model, I intend to focus on Positive Reinforcement. Positive Reinforcement involves presenting a good consequence (reward) when your pet performs the desired response. The reward can be a treat, a toy, warm affection, or praise. Your dog may perceive each one of these as a good consequence for his desirable behavior. For instance, if you say “sit”, and your dog sits, you then give him a treat. The treat is a reinforcer which is intended to increase the likelihood of future desirable responses.

Rewards (positive reinforcement) applied in a timely fashion will increase the likelihood that your dog will respond each time you present the command. Positive reinforcement strengthens behavior by teaching your dog that there are consequences for his behavior. Numerous studies have proven that dogs are able to make associations within one to two seconds.

This is one of the most critical principles for any dog trainer. What that means for you as the “trainer” is that you must reward your dog immediately following the compliant response. Timing is of the utmost importance! If your reward presentation is delayed, you may, unknowingly, be rewarding a completely different behavior. For example, if your dog sits to your command and you delay the reward for 5 to 6 seconds, your dog may have sniffed the ground or scratched his ear after sitting down. You are rewarding him for sitting, but he may think you are rewarding him for sniffing the ground or scratching his ear. In this scenario you may discover that your sit command may result in a “sit” and “scratch the ear” response. An undesirable association occurred when the reward was delayed after the dog obeyed your “sit” command, which was not at all your pet’s fault.

In recent years, Positive Reinforcement has become the politically correct way of gaining desired results. People tend to think that you can train a dog by using the positive reinforcement model alone. If it is used inappropriately, the results will likely turn out negative, causing your pet some confusion. You may never be able to inhibit a bad habit by using positive reinforcement. It is contradictory to scientifically proven learning models, as well as, nature itself. 

For example: Let’s say that your dog is a persistent jumper. His way of greeting people is by jumping up onto them. Perhaps you have heard or read somewhere that if your dog jumps on you, you should respond by turning to the side and ignoring him until he sits. Once he sits you then give him his reward or treat. It is assumed that in time he will learn that he is rewarded for sitting and the undesirable jumping behavior will stop. Once you understand the anatomy of Positive Reinforcement, you will clearly see that this kind of advice goes against what has already been proven through extensive studies. In fact, in reality you are rewarding the dog for jumping. The dog has learned that it first must jump, then, sit, in order to be rewarded. Rather than inhibiting the jumping behavior, this technique reinforces it. This would be a prime example of the inappropriate use of the Positive Reinforcement model.  

Keep in mind that Positive Reinforcement is only one piece of the four part learning model. Beginning with one part rather than the whole will undoubtedly lead to error.  Effective and balanced dog training encompasses teaching principles associated with the entire learning theory.

In the next segment, the Negative Reinforcement model will be addressed. A working description and effective application will be the primary focus.
 

Negative Reinforcement

 In our last article we discussed Positive Reinforcement which is one part of a four part learning model developed by B.F. Skinner which is known as Operant Conditioning.

Negative Reinforcement is misunderstood by a lot of people due to the fact that it is perceived by many as punishment. As you will see, it is an effective learning principle when applied properly to teach your dog.

The word reinforcement actually means to strengthen. When a behavior has been made stronger, it means each time you ask your dog to perform the desired behavior; he will be more likely to comply with the request. Let me give you an example using a horse. Let’s say that you are riding along a trail and you want the horse to turn to the right. You would pull on the right rein which would put a little pressure on the bit in the horse’s mouth. The horse has learned from repetition when he yields to the slight pressure, the pressure goes away. Yielding to the pressure becomes self-rewarding for the horse which results in the horse learning to respond to the physical cues given through the bit in his mouth.

The same cues can be given to a dog to help a dog to learn what is expected of him. One example might include pushing dog on a dogs back side in order to teach the dog to sit. The pressure on the dogs back side continues until the dog sits. As soon as the dog sits, the pressure is removed. The pressure or physical cues is something that is slightly unpleasant which is removed immediately when the dog complies with the request.

Some trainers refer to this technique as “escape training.” The dog learns that he can escape the “slightly unpleasant physical cue” by complying with the request. When your dog is taught how to escape the cue, it becomes self rewarding, because the dog wants to remove the cue as fast as possible.

As with a bit in a horse’s mouth, so is it with a collar on a dog’s neck. Every piece of training equipment communicates something to the animal. Some collars may be very effective in communicating to one dog, but inappropriate for another dog. As a result, some training collars are shunned while others are recommended as a quick fix for any dog.

In our next article we will consider punishment. It’s not a subject that most people want to talk about, but it is one aspect Operant Conditioning. It is sometimes confused with Negative Reinforcement. Learning theories can be complicated when certain parts are over emphasized while other parts are ignored. The purpose of this four part series is to help empower you through a more thorough understanding of all four aspects of Operant Conditioning.

Negative Punishment

Parts 1 and 2 of this series examine the fundamental characteristics of B.F. Skinner’s operant conditioning theory. These two segments discuss positive and negative reinforcement. If you would like to review them, you may visit my website at: www.smartdogtrainer.com and click on Articles.

Punishment is not a word that you hear too often with regards to dog training. One definition of punishment; severe, rough, or disastrous treatment prompts people’s minds to conjure up all kinds of abusive ideas. This especially applies to animals. However, Skinner defined punishment as a means of inhibiting (stopping) some kind of unwanted behavior. Although most people consider punishment as negative, Skinner uses the term “negative punishment” to indicate that a reward is being negated, subtracted or removed from the training sequence. In other words, by withholding a reward when your dog fails to comply with your command “the good things in life” go away.

Here is an example: Let us assume that every puppy is motivated by “treats.” Your puppy can be trained to perform the sit position by being lured by the treat. Immediately at the moment when his behind touches the floor you must deliver the reward (treat). If you perform this luring technique over and over your puppy will catch on rather quick. After 15 minutes he will sit every time upon your command. He may “sit” seven times in a row, yet on the eighth request your puppy’s response is to just stand there and ignore you. Would you give him a treat? No! You would withhold the reward because he failed to comply with your command.  This is what B.F. Skinner meant by negative punishment.

Remember that “negative punishment” is a way of inhibiting (stopping) an undesirable behavior. In this example we withhold the reward because we want to stop the dog’s resistance to the “sit” command. The puppy will learn (become conditioned to) that if he doesn’t obey, he doesn’t get rewarded. But, what if your puppy isn’t motivated by treats at all? What if your puppy has such a strong chase drive that it runs after anything that is moving? Well, I guess that negative punishment doesn’t work with your puppy. You may have a child who doesn’t respond to negative punishment either.

Perhaps you are waiting for a profound answer to these frustrating questions. The fact of the matter is, operant conditioning is a Four Part Learning Model.  There is nothing more likely to lead to error and frustration than to begin with a part rather than the whole. The key to negative punishment is “desire.” What is it that really captures your puppy’s attention the most? In the next article we will examine the fourth and final part of B.F. Skinner’s theory of operant conditioning known as “positive punishment.”

The Anatomy of Positive Punishment

 This is the last of a four part series describing the four parts of B. F. Skinner’s theory of Operant Conditioning. Punishment is subject that has been ignored for the most part in contemporary thinking because it is considered to be archaic. More often than not punishment is equivalent to abuse. However, when defined in the context of Skinner’s theory of operant conditioning, it can be viewed as an effective way of inhibiting a bad behavior.

 Positive punishment is some bad consequence which is brought to the training sequence as a result of a wrong response. That is to say that when presenting a bad consequence, the behavior is less likely to occur. Skinner said,”If punishment is to be effective, it must be severe enough to inhibit the behavior.” Let me give you and example. Let’s say that you are traveling down the expressway at 90 miles an hour and you are pulled over by a New York State Trooper. You are given a speeding ticket and required to report to the court to pay the consequences for you bad driving behavior.

 When the judge reviews the evidence, he decides that the penalty for driving 90 miles an hour, 25 miles over the speed limit is only $2.00. Would a $2.00 fine change your driving habits? I don’t think so. Let’s say that the judge decided that the fine ought to be $10,000, and revokes your driver’s license?  Would that level of penalty change your driving habits? After a penalty like that, you may never want to get behind the wheel of a car again. Obviously, neither penalty is adequate. I have described the two extremes which have both contributed to the confusion surrounding the use of punishment as a means of inhibiting a behavior.

 For those who don’t understand punishment, this aspect of operant conditioning has become increasingly more difficult to implement. We would like to be able to say that on a scale of one to ten that every dog will stop an annoying behavior with a level 3 correction. If that were the case, training dogs would be easy. However, there are circumstances that change like the weather, and what constitutes punishment for one dog, might be rough play for another dog. So how do you know what level of discipline is enough (unpleasantness) and what is too much? It helps to understand if your dog is “soft or hard.”

 A soft dog is the sensitive type, while a hard dog is anything but sensitive. The softer dog requires much less discipline to communicate undesirable behavior while the harder dog may need a firmer hand. Depending upon the behavior, it is recommended that you consult a professional when dealing with long standing problems. However, there are some general guidelines. First, if you find that whatever you may be doing to stop a bad behavior doesn’t seem to be working, then chances are that what you are doing isn’t severe enough to actually inhibit the behavior. On the other hand, if you find that your dog is becoming fearful, then you are using too much force.

 As with all of the four different parts of operant conditioning, timing is critical to the learning process. If you are going to reward a dog for doing something right or punish a dog for doing something wrong, you have only one and a half seconds to do it if your dog is going to make the right association. For punishment to be effective, you must catch the dog in the act of the behavior that you are trying to inhibit. If too much time elapses before the punishment is administered, your dog will not know why or what you are disciplining him for. Poor timing will confuse your dog and is frankly, abuse.

 There is nothing more likely to lead to error than to begin with the part rather than the whole. Operant conditioning is a four part learning model and should be applied in such a manner as to employ all four parts to the learning process. To emphasize one part and to ignore another part will only result in a confused dog and a frustrated owner.

Please keep in mind that all four parts of this learning model work synergistically of one another, NOT independently. They need to be elastic and flexible depending upon your unique circumstances.

Contact Information

Tom Beitz is the owner of the Academy for Puppies and Dogs and is an authorized dealer for Pet STOP Hidden Dog Fences. Tom can be reached at (716) 628-0651 to answer your questions or he can be found on the web at www.smartdogtrainer.com . E-Mail: Tom@Smartdogtrainer.com

Copyright © Academy for Puppies and Dogs 2007 (All Rights Reserved)

Copies of this article may be freely distributed provided the original source is cited!